ras genes are essential for cellular growth control in eucaryotes. Moreover, activated ras genes are among the most commonly found oncogenes in human tumors. In the yeast Saccharomyces cerevisiae, an activated ras allele (analogous to oncogenic mutations in mammalian ras) causes the overstimulation of adenylyl cyclase. Elevated cyclase activity results in uncontrolled cell proliferation, an inability to arrest in response to nutrient starvation and acute sensitivity to heat shock. Despite great strides in understanding the biochemistry of ras in yeast, the ras pathway in mammalian cells differs and remains unclear. This proposal describes the isolation, by selecting for the ability to suppress heat shock sensitivity in yeast cells harboring an activated ras allele, of mammalian cDNAs that encode ras-interacting proteins. Our initial cDNA isolates include phosphodiesterases that 'suppress' by lowering intracellular cAMP levels. Three additional clones from our mammalian cDNA expression library apparently encode proteins that interfere with ras function through a direct interaction with ras (both wild type and oncogenic). These are likely to suppress ras function either by decreasing the proportion of active ras in the cell (down regulation), or by forming a nonproductive complex with ras and lowering its effective concentration (dominant negative interference). The latter class should include truncated ras effectors that bind to ras but are unable to transmit growth signals. All of these isolates were partial cDNAs. To ascertain the normal functions of these gene products, full length cDNAs will be isolated and expressed in our model system. The minimum portion required for interference with ras function will also be determined by deletion mapping. A central focus of this work involves cell culture experiments to determine the role of the normal gene products in mammalian cells. If they normally act as ras suppressors or down regulators, then overproduction should interfere with ras function in mammalian cells, as judged by their ability to either block or reverse ras transformation. If our full length clones encode effector proteins, then their overexpression in mammalian cells may elevate, rather than antagonize, ras function. Effects on ras-induced differentiation will also be examined. The results of these studies should greatly expand our knowledge of the function of ras in mammalian cells. They should also lead to a better understanding of oncogenesis in general.